Re: WARNING! New Perl/Perl-base upgrade removes 141 Sid/Unstable packages
On Fri 16 Dec 2016 at 12:40:29 (+0100), Vincent Lefevre wrote:
> On 2016-12-07 23:45:24 +0000, Lisi Reisz wrote:
> > On Wednesday 07 December 2016 14:55:40 Vincent Lefevre wrote:
> > > On 2016-10-13 00:09:02 +0100, Lisi Reisz wrote:
> > > > On Friday 07 October 2016 15:43:17 Vincent Lefevre wrote:
> > > > > On 2016-10-04 22:51:34 +0100, Lisi Reisz wrote:
> > > > > > On Tuesday 04 October 2016 08:25:46 Vincent Lefevre wrote:
> > > > > > > my position remains the same:
> > > > > > > aptitude is poorly designed.
> > > > > >
> > > > > > Fine. So don't use it. But moaning won't help anyone, not even you.
> > > > > > You don't like Aptitude. We get the message. So don't use Aptitude.
> > > > >
> > > > > And what do you propose instead?
> > > >
> > > > I don't use Sid, so haven't tested out which package managers are good
> > > > for it when there are problems, but how about looking at apt or apt-get?
> > > > Ben says that he has great success with apt-get. Apt-get is much less
> > > > aggressive than aptitude - but less fully featured.
> > > >
> > > > If I use aptitude with a large number of upgrades, I try to break it up.
> > > > At the very least I do
> > > > # aptitude update
> > > > #aptitude -s safe-upgrade
> > > > # aptitude safe-upgrade
> > > > # aptitude -s full-upgrade
> > > > # aptitude full-upgrade
> > >
> > > Sorry for the late reply, but all these may remove important packages,
> > > i.e. they have the same issues.
> >
> > Don't let them - that is the point of the -s.
>
> The -s is actually not necessary since if a package is to be removed,
> the user may still refuse.
It protects one from making a mistake by a lapse of attention while
focussing on the problem in hand, ie examining lists of packages that
apt* wants to remove. Why discourage its use?
But more than that, -s allows one to
$ aptitude -s full-upgrade
rather than
# aptitude -s full-upgrade
so one can investigate these issues without even taking on root
privilege at all.
Cheers,
David.
Reply to: